Application of Pitzer’s Equations for Modeling the Aqueous Thermodynamics … · Aqueous Thermodynamics of Actinide Species in Natural Waters: A Review~ n Andrew R. Felmy* and Dhanpat

, .. 1

Journal of So!ution Chemistq, Vol. 28, No. 5, 1999

Application of Pitzer’s Equations for Modeling theAqueous Thermodynamics of Actinide Species inNatural Waters: A Review~

nAndrew R. Felmy* and Dhanpat Rai

Received October 30, 1998; revised February 2, 1999

A review of the applicability of Pitzer’s equations to the aqueous thermodynamicsof actinide species in natural waters is presented. This review includes a briefhistorical perspective on the application of Pitzer’s equations to actinides, informa-tion on the difficulties and complexities of studying and modeling the differentactinide oxidation states, and a discussionof the use of chemical analogsfordifferent actinide oxidation states. included are tables of Pitzer ion–interactionparameters and associated standard state equilibrium constants for each actinideoxidation state. These data allow the modeling of the aqueous thermodynamicsof different actinide oxidation states to high ionic strength.

KEY WORDS: Actinides; tkrnmdynamic dat& Pitzer ion-interaction pammete~,trivatent actinidcs tetmvalent actinids, pentavalent acdnidrw hexavalent acrinides.

1. INTRODUCTION

Actinide speeies in aqueous solution can exist in a variety of oxidationstates (III, IV, V, and VI) and can strongly interact with several differentIigands of importance in natural waters such as OH-, CO:-, SO%-, and F-.This large combination of oxidation states and interacting Iigands makes itnecessary for aqueous thermodynamic models to include large numbers ofchemical species and treat a wide range of electrolyte types from simple 1:1electrolytes to highly charged 4:2 and even 6:1 electrolytes. In addition,

Pacific Northwest National Laboratory, Battelle Boulevard, PO Box 999, Rtchland, Witshing-ton 99352.twe d~mte ~i5pawrtohe memo~ of Kenneth S. Pitzer in rWO’#titiOtS of his manY

invaluable contributions to solution chemistry.

533

{XW5-9782YYW5(KK1533S i 6.(IM Q IVYY Plenum Fubiisbhg Cmpmation

DISCLAIMER

This repoti was prepared as an account of work sponsoredbyanagency of the United States Government. Neither theUnited States Government nor any agency thereof, nor anyof their employees, make any warranty, express or implied,or assumes any legal liability or responsibility for theaccuracy, completeness, or usefulness of any information,apparatus, product, or process disclosed, or represents thatits use would not infringe privately owned rights. Referenceherein to any specific commercial product, process, orservice by trade name, trademark, manufacturer, orotherwise does not necessarily constitute or imply itsendorsement, recommendation, or favoring by the UnitedStates Government or any agency thereof. The views andopinions of authors expressed herein do not necessarilystate or reflect those of the United States Government orany agency thereof.

DISCLAIMER

Portions of this document may be illegiblein electronic image products. Images areproduced from the best available originaldocument.

c. .

534 Felmy and Rai

since many actinide compounds are insoluble, the chemical species of theseelements are usually present in solution at trace concentrations relative tothe major electrolyte components (e.g., Na+, Cl-, CO~-, . ..). This meansthere are always minimums of at least three chemical species present insolution. Within the Pitzer activity coefficient formalism, the activity coeffi-cient for a trace species in these systems’ ..

can involve several model parameters including mixing terms for ions of likesign (i.e., (3and W) as well as mathematically redundant cation-anion terms(i.e., ~(o)in the B,x term in Eq. 1) between the tracer cation or anion and theoppositely charged bulk cation or anion. Such facts make it difficult to defineunambiguous or nonredundant electrolyte ion-interaction parameters for traceactinide species when such parameters often must be fit from common-ionternary data. In addition to these complications, the necessity to treat highlycharged electrolyte species presents a particular challenge both in terms ofthe magnitude of the higher order electrostatic terms in unsymmetrical mix-tures and in terms of model parameterizations involving the ~(z) term origi-nally proposed by Pitzer in the late 1970s.(3’4)

For these reasons, application of Pitzer’s equations to the study of theaqueous thermodynamics of actinide species has presented a stringent and

challenging test of the theory. Such applications are important given the needto understand and predict the aqueous thermodynamics of actinide speciesin nuclear waste disposal areas.

The remainder of this paper will begin with a brief historical perspectiveof the development and application of Pitzer’s equations to actirtide aqueousspecies followed by a detailed summary of the Pbsr ion–interaction parametersfor actinide species of importance in natural waters. Thk summary will bepresented for different oxidation states beginning with the trivalent actinides andending with the hexavalent actinides. Patarneter vahtes will be included for themajor ligands of importance in natural waters including OH-, CO;–, S@4-, F-,

.. and W4-, as well as for selected organic Iigands (i e., EDTA). Tables of modelingpararnetets applicable to each oxidation state will also be presented.

*All of the terms in Eq. (1) are defined elsewhere.(12] Briefly, t represents the tracer cation, Band C are second and third viriat coefficients specific for each cation–anion interaction, @is a second virial coefficient specific for cation-cation or anion-anion interactions and isnumerically equal to 13ijplus ‘(jij. Oijis a constant model parameter for each cation-cation oranion-anion interaction. %)ijis the higher order electrostatic term for unsymmetrical mixtures.F is a modified Debye-Huckel activity coefficient term, z is the species charge, Z equalsZlzilmi, and WMXis a constant third virial coeftlcient representing cation-cation-anion oranion-anion-cation interactions.

.

Aqueous Thermodynamics of Actinide Species 535

2. HISTORICAL PERSPECTIVE OF THE DEVELOPMENT OFPITZER ION-INTERACTION PARAMETERS FORACTINIDES

The first work on modeling the aqueous solution thermodynamics of actin-ide species using Pitzer’s equations was done by Pitzer and Mayorga.[s) Thisoriginal work included model parameters for hexavalent uranyl(UO~+) with Cl-, C1O;, and NO;, and tetravalent Th4+ with Cl- and NO;. Inaddition, model parameters were developed for several trivalent rare-earthcations with Cl-. One year later, Pitzer and Mayorga@ published the parametersfor the 2:2 electrolyte UO~+–SOj-. These initial studies largely completed theparameterization of the few binary chemical systems where the actinides oractinide analogs (such as trivalent rare-earths for trivalent actinides) are suffi-ciently soluble to allow the determination of unambiguous cation–anion inter-action parameters. The vast majority of subsequent work has focused onchemical systems where the actinides (An) species are present in trace concen-tration relative to the major electrolyte solution components.

The first of these trace activity coefficient studies was conducted in1989 on the solubiiity of PU(OH)3(S) in complex brines!’) In this application,ion–interaction parameters for Nd 3+–C1- were successfully used as analogsfor the unknown and difficult to experimentally measure (because of oxidationproblems) I?u3+-C1- ion–interaction parameters. These studies establishedthe usefulness of trivalent analogs in estimating the Pitzer ion–interactionparameters for diflicult to measure actinide species. The use of these parame-ters accurately described the two to three orders of magnitude increase insolubility of PU(OH)3(S) in concentrated brine over that observed for diiutesolution (Fig. 1). These studies established that the use of the Pitzer equationsoffered a simpler approach, over that of assuming the formation of multiplePU3+–CI- complexes, and accurate formalism for modeling the volubility ofactinide species in complex brines. The success of this work resulted in theearly 1990s application of Pitzer’s equations to a wide range of chemicalsystems involving actinide species. Much of this work was performed at thePacific Northwest National Laboratory (PNNL) for trivalent and tetravalentactinides$7-24) the Lawrence Berkeley National Laboratory (LBL), SandiaNational Laboratory (SNL), Los Alamos National Laboratory (LANL) forpentavalent species, principally Np(V)@-2*), and at the Institut fur Radio-chemie in Karlsruhe Germany for trivalent and pentavalent actinides@-33).These studies, combined with previous work by Pitzer, resulted in the develop-ment of a significant database of modeling parameters for the chemical systemAn(III)–Cl–SOi–P04–C03–Mo04–Hz0 at 25°C, for the system Th(IV)-Cl–F–S04–C03–H20 at 25”C, and for pentavalent species, principally Np(V).In addiiion, development of Pitzer ion–interaction parameters for organic

,

536 Feimy and F&d

-3 I I

-4

-s

-6

-7

-8

-9r

-10t

\

\\‘d-

\

Pu(OH)Jam) “1+ Pa

indiluteNaCl+*

h btha

\

iC@ec$ontimit

------- ------ - ‘+-------------

I-111 1 1 ! I ! J

o 2 4 6 .8 10 12 14

Pc”+Fig. 1. The effect of PBB I brine ([N-a+] = 5.2 M, [Ca2+] = 0.034 M, [Cl’] = 4.6 M,[SO%-] = 0.W4 M, ionic strength - 5.5 M) on the volubility of PU(OH)l (am). From Felmyet al., Ref. 7.

cttelate-actinide species has been undertaken recently at Florida State Univer-sity and Sandia National Laboratory.c~J Other relevant studies involvingtrivalent rare-earths have been previously conducted at the University ofMiami with applications to seawater systems. These actinide studies inconjunction with binary and ternary ion–interaction parameters for majorelectrolyte ions (Appendix Tables A.I and A.II) are providing a useful andreliable database for application to complex natura[ systems containing actin-ides. Specific aspects of the development of this database are discussed belowfor each actinide oxidation state.

3. ION-INTERACTION PARAMETERS FOR DIFFERENTACTINIDE SPECIES

3.1. Trivalent Actinides

Of all of the actinide oxidation states, trivalent actinides, and rare-earthshave been the most extensively modeled using the Pitzer formalism. In fact,with relatively few modeling parameters (Tables I and II), this combinedspeciation-ion-interaction approach has successful y modeled the availablevolubility and other thermodynamic data for an extensive chemical system

.

.

.

. ..’;.

,, .,

Aqueous Tbermodynamies of Actinide Species 537

Table f. Binary and Ternary Ion–Interaction Parameters for An (111)Species

Binary parameters,

Species ($0) pm p(z)d’ Ref.

AS$+-C}- 0.5856 5.60 0.0 –0.0166 31AnClz+-Cl- 0.593 3.15 0.0 –0.006 31AnClj-Cl- .516 I .75 0.0 .01 31AnOH2’–CI- –.035 1.6 0.0 .05 31An(OH)~-Cl- -.616 –.45 0.0 .05 31

. An3+-SOj- 3.0398 0.00 –2500 0.00 13An3+-H2PO: 0.00 0.00 –92.9 0.00 0

Na+-An(CO& –0.256 5.0 0.0 0.044 b

Ternary parameters Ref.

Nit+-An(OH)Jaq) –0.2 31C1--An(OH)Jaq) –0.2 31Na+-C1--An3+ 0.1 31Ca~+_C1--An~+ 0.2 31Ca~+-Cl--An@+ –0.014 31Caz+–Cl--AnCl~ –0.196 31Cl--An(COJ~- 0.168 c

Na+-Cl--An(COJ]- 0.0273. c

u ASSu~ to & ~nakgous to the parameter for N&l+–HzPO~ (Ref. 10).

bAssumrxI to be analogous to the parameter for Na+–Nd(COJj - (Ref. 14).cAssumed to be analogous to the parameter for Na+-Cl ‘– Nd(COJ~- (Ref. 15).

Table II. Log R Values (25”C) for the Formation of An(lII) Aqueous Species

Reaction @ ~ Ref.

An’+ + HZO = An(OH)z+ + H+ –7.56 31Ans’ + 2H20 = An(OH)~ -i- 2H+ – 15.8 31Ar$+ + 3HZ0 = An(OH)$ + 3H+ <–28.6 8An~+ + CI- = An@+ .24 31Ar$+ + 2CI- = AnCl~An3+ + CO;- = AnCO.~

-.74 317.6 .8 ..

An3+ -t- 2CO~- = An(COJ); 12.3 8An3+ + 3CO~- = An(COJ$- 15.2 8An3+ + 2MoO~- = An(MoO& 11.2 a

. 2Am’+ + M@02~OH)j- = AnzMo@Jaq) + 4H+ 3.85 a

Ar$+ ‘+ F- = AnFz+ 3.4 38An~’ + 2F- = AnF~ 5.8 38An3+ + 3F = AnFT(aq) <11.2 b

“Assumed to be analogous to the values for corresponding Nd(HI)/molybdate species. (Ref. 9).hEstimated from the results for volubility of NdF3(c). (Ref. 12).

I

.

538 Felmy and I&ii

(i.e., An(III)-Cl-S04-POA-COJ-Mo04–HZ0 at 25”C) to high ionic strength.The vast majority of this model development has focused on the rare-earths,Am(III), or Cm(III). Very little work has been done with Pu(HI) because ofexperimental difficulties in maintaining Pu as Pu(III). However, the limitedstudies that have been done [i.e., PU(OH)3(S) in brines], along with the knownhydrolysis constants and soh.tbility products for Pu(IH) species, support the .use of modeling parameters developed from Am(III)/Cm(III)/trivalent rare-earths for Pu(HI). As an example, the sohibility data for the hydroxides ofAm(III), Pu(III), and Nd(III) are all similar (Fig. 2), resulting in similar ,volubility products and hydrolysis constant<lon–interaction parameters.

As previously described, the development of the Pitzer ion-interactionparameters for trivalent actinides has been conducted at PNNL using rare-earth analogs and at the Institut fiurRadiochemie using Cm(III) isotopes. Thedevelopment of the model parameters by both groups is consistent in thatsimilar sets of major electrol yte parameters (see Appendix) were used in the . .

model calculations. Table I presents the Pitzer ion–interaction parameters forAn(III) species and Table II the standard state. equilibrium constants. In somecases (Ci-, F– and MoO%- parameters), the seiected values were the ordyvalues in the literature that had been extensively compared to a wide rangeof experimental data. In other cases (OH-, CO~- and SO~-), more than oneset of model parameters satisfied this criteria.

-1

-3

-s

-7

-9

-11

.4

● Nd(OH)Jc)mm Nrt(oH)3(cp’)+ Am(OH),(c)Wo Adsv Pu(OH)3(am)m

o 2 4 6 8 10 12 14

pc”+Fig. 2. The S.oiubltityof trivalent actinide aud Ianthanide hydroxides in dilute solutiow (Ref. 39).

—

. .

, .


The hydrolysis constants and ion–interaction parameters for hydrolysisspecies were taken from the recent compilation of Fanghanel and KimJ3’JThese data were selected since this is the only set of data with ion–interactionparameters between the An(III) hydrolysis species and the major electrolytes(at least Na+ and Cl-) and the parameter values have been tested to somedegree against independent experimental data [i.e., the volubility ofAm(OH)J(s) in NaCl]. The hydrolysis constants are also in reasonable agree-ment with other values in the literature.@5J

The parameters for carbonate and sulfate were more difilcult to selectsince two different sets of model parameters are in the literature. The firstset was developed at PNNL by the authors of this paper and the second sethas been developed at the Institut fur Radiochemie. The data set developedat the -Institut fiir Radiochemie contains significantly more modeling parame-ters than the PNNL data set, principally as a result of the emphasis by theInstitut fur Radiochemie group in explaining their time-resolved fluorescencespectroscopy (TRLFS) measurements of Cm(III). The PNNL model parame-ters have also been compared to a significantly wider range of experimentaldata and it is, therefor~, of interest to test the model parameters of the Institutfiir Radiochemie group on some of these data sets. As part of this effort, wehave initiated this process.

In the case of carbonate, we have compared the Institut fiir Radiochemiemodel parameters to the experimental data of Rao et af.(14’15)on the volubilityof NaNd(C03)2 “ 6H20 sodium carbonate, sodium bicarbonate, and in mixedelectrolytes with NaC1. On an overall basis, the parameter set proposed bythe Institut fur Radiochemie group works quite well on this extensive set ofindependent data. However, a couple of facts prevented us from selectingtheir parameter set. Fkst, although their model gives quite satisfactory agree-ment with the NaNd(COJ2 “ 6H20 volubility data in NaHC03 solutions, thecalculated speciation does not appear reasonable in that An(C03); andAnCOJ are the dominant species in 0.1 m NaHC03, but uncompleted An3+is the dominant species in 1.Om NaHC0,3. In addhion, the model parametersproposed by the Institut fur Radioehemie group slightly overpredict (-0.5

., log units) the observed NaNd(C03)z “ 6H20 solubilities in NaCl + Na2C03 <solutions at high NaCl (i.e., 4m).

In the case of sulfate, we have compared the Institut fiir Radiochemiemodel parameters to the experimental data of Rai et uf.(’3) on the volubilityof NdP04 in Na2S04. The model parameters from the Institut fur Radlochemiegroup work quite well for the dilute (-0.001 m sulfate) solution, but underpre-dict the volubility data at higher sulfate (-O. 1 m) by approximately and orderof magnitude. Although we cannot be absolutely sure, this underpredictionappears to be the result of a relatively low value for the stability constant ofthe An(S04); complex (see Rai er aL(’3) for a review). Although further tests

..

, .

,,

. .

.>...*:

‘>.>.,. .,.-.. /

540 Felmy and Rai

of the different modeling approaches would be valuable, both data sets repro-duce a wide range of experimental data quite satisfactorily and this shouldgive users of these models confidence in using either approach.

This database of modeling parameters makes application of Pitzer’sapproach to predicting the aqueous thermodynamics of trivalent actinides in

natural waters to high ionic strengths a reliable possibility. This is clearlydemonstrated by the data and calculations (Figs. 3–5), which show the appli-cability of the model to concentrated Na2COs and NaHC03 solutions (Fig.3), the predictions in solutions containing high concentrations of polymerizedmolybdate species (Fig. 4), and the applicability to concentrated Na2S04solutions (Fig. 5).

3.2. Tetravalent Actinides

The vast majority of Pitzer modeling parameters for tetravalent actinidesis available for Th(IV). Th(IV) is the easiest tetravalent actinide to studyexperimentally because the tetravalent state is the onl y oxidation state of Ththat is stable in aqueous solution. For example, modeling parameters for theTh–carbonate system were developed based on extensive experimental dataon the volubility of Th02(am) as a function of H+, CO:–, and HCOI concentra-tions (Fig. 6). The direct use of Th(IV) model parameters for U(IV), Np(IV),and Pu(IV) species is expected to be problematic, because of systematic and

-20 f 1

-Em

-3.0

4.0

-5.0

-6.0

-7.0

EEsl

.

,

.

1.-6.0 \ 1 I ! t I I

.,

.. ...

,-

,. ..:’:,...

;“,

, :!

“:,,

‘.

-12 -1.0 4.8 -0.6 4.4 -0.2 0.0 02 0.4

log (Cti c /mol din-’)

Fig. 3. Observed and predtcted soiubility of NaNd(COJz(s) in concentrated NazC03 andNaHC03 solutions (Ref. 14).

..

Aqueous ‘1’bermodynamics of Actinide Speeies 541

. . .

4

-6 --

● ●

✍✍✍✍✍✍✍✍✍✍✍✍✍✍✍✍✍●✍✎ I

\

#--8 .. .“---

Cywlafti “f::wJMledialkra (noMOOJ * -

_____ -------- - -------- ----- ------ --- ----- -

-9 t ! I I I I 1

1 2 3 4 5 6 7

pc”+Fig. 4. Volubility of NdPOi(c) as functions of PCH+ and molybdate concentrations at fixed

[phosphate],(Halof 10-’5 M (Ref. 9).

large differences in charge to ionic radii, and other factors, for the differenttetravalent actinides. Nevertheless, for the few cases where data do exist forTh(IV) and other tetravalent actinide complexes other than hydroxide [i.e.,U(IV)], the Pitzer ion–interaction parameters for Th(IV) species have provedto be reliable indicators of the expected magnitude of the parameters for theother tetravalent actinides [e. g., compare the Th(IV) data in Tables III andIV with U(IV) data in Table V]. Values for the Pitzer ion–interaction parame-ters for Th(IV) species with Cl-, SO~-, and CO:- are available at 25°C aswell as the necessary equi Iibnum constants for these and other species (Tables111and IV). This model development effort represents a significant test of .the Pitzer ion–interaction model given the potential number of chemicalspecies present and the high charge type for many of these ion interactions.In particular, published values for the Th(COJ$- complex represent the onlyaccurate values for this highly charged (6: 1) electrolyte type present in theliterature. Also note that when highly charged species are present in solutions,the magnitude of the mixing terms (i.e., 6 and W) for such species with thebulk cations or anions can be very large and important in predicting theaqueous thermodynamics. For example, Felm y et aL(’9) have estimated thevalue of 0 for Th(C03)~-–C10~ at 5.5. Such large mixing terms for highlycharged species means that models that attempt to describe the thermodynamic

I

.

Felmy and Rai

— Predicted using $0)= 3.0396 and pm = -2500 forflrn=- *

-4 -3 -2 -1

tog304 (moi ● L-l)S’x?mlas

Fig. 5. Normalized equilibrium aqueous Am concentrations from solvent extraction data ofMcDowell and Coleman, (Ref. 42), obtained at constant acid activity and extractant-phasecomposition. The solid line represents the predicted concentrations (Ref. 13).

.

relations for such species based solely on binary (cation-anion) interactions(such as, the S.I.T, modelt29)) will fail for highly charged electrolyte specieswhen applied to common-ion ternary or other multicomponent solutions. Thisfact was discussed in detail by Felmy et d.(19) in describing the differencesin Th02(am) volubility in solutions with and without added NaCIOd. Furtherevidence for this is demonstrated in Fig. 7, which shows experimental datafor Th02(am) in mixed NaCl and Na2C03 solutions are drastically differentthan model predictions based solely on binary data [i.e., onlyTh(COJ$--Na+ interactions determined from NazC03 solutions] Qr data inNaCIOg [i-e., setting $ for Th(COJ$--CIO; to 5.5]. Both approximations .are in error by orders of magnitude, although the actual calculated parametersfrom fits to these experimental data are in the same range of expectedmagnitude [i.e., 0 for Th(COJ~-–Ci - and W for Th(COJ$--Cl--Na+ at 1.8 .and 0.3, respectively]. Clearly, highly charged electrolytes interact stronglywith their ionic environment and this ionic environment for NaCl solutionsdiffers from NaCIOA or other solutions. These facts necessitate the inclusionof mixing terms in the aqueous thermodynamic models. Such situationspresent difficult challenges for any electrolyte model, challenges that can be

..... . --------—... -_.—_-_:_____

Aqueous Thermodynamic of Actinide Species 543

.

.

.

‘O 0.4 0.8 1.2 1.6 2.0N~C03 (M)

t — CeIculated

-2 :(l-hisstudy)

● Experimental

-3 ~

-4 + ●

+ ;0

-7 =

.~-1.6 -1.4 -1.2 -1.0 -0.8 -0.6

Log HCOa(M)-2 -

— CeIculeted

-3 :(TldSstudy)

. Experimental#

-4 –

-5 ~

-8- I I , I , I I I t I

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0

Log NaOH (M)

. .

..

. . ...; .

.

..

w Felmy and Rai

Table 111. Relevant Pitzer Ion-Interaction Parameters for Th(lV) Species

Binary parameters

Species p(o) p(l) $(2) C+ Ref.

H+-Th(SOJ~- 0.84 0.00 0.00 0.00 16Na+-Th(SO.J~- 0.12 0.00 0.00 0.00 16 .K+-Th(SOq)~- 0.90 0.00 0.00 0.00 16Th4+-cl- 1.092 13.7 –160 –0.112 46

Th4+-HSOZ 1.44 0.00 0.00 0.00 16Th4+-so;- 1.56 0.00 0.00 0.00 16Na+–Th(COJj- 1.31 30.0 0.00 0.00 19 ,


H+_Th4+ 0.60 16H+-Th4+_(3- 0.08 16Na#-~g+ 0.42 20Na+–Thg+_Ci- 0.21 20Mu2+-Th4+ 0.60 20&+-~4+_CI- 0.21 20Th(so4)2-cl- 0.29 16Th(SO,)z-HSO: 0.68 16

Table IV. Log P (25”C) for Reactions Involving Solution Species of Th(IV).

Reaction Log r Ref.

Tif+ + 2s0?- = Th(so&(aq) 11.59 16Th4+ + 4H20 = Tag + 4H+ =—19.7 47Ti#+ + 3s0;- = Th(so4):- 12.42 16Th4+ + 3F- = ThF.: 18.89 18Th4+ + 4F- = ThF4(aq) 22.33 18Th4+ + SF- = ThFI 24.76 18Th4+ + 6F- = ThF~- 25.56 18

. 4

..

handled by the Pitzer approach, but cannot be handled by simpler approaches(e.g., S.I.T) that include only cation–anion interactions.

In addition to the data available for Th(IV) (Tables III and IV), significant .

Fig. 6. Experimental data of Rai et al. (Ref. 43) and predictions line based on the model ofFelmy er al. (Ref 19). Aqueous Th concentrations in equilibrium with ThOz(am) in NaZC03solutions containing O.IM NaOH (top), NaHC03 solutions (middle), and NaOH solutionscontaining 1.0 M NazC03 (bottom).

. h


Table V. Pitzer Ion–lntemction Parameters and Log ~ (25”C) for U(lV) Species

Reaction L@g~ Ref.

U4++HZO= UOH3+ +H+ –0.50 48

L.P+ +So%-=uso;+ 9.0 21U4+ -t- 2SO~- = U(S04)z(aq)U4+ + 5CO:- = u(coJg-

11.7 2131.29 22

LP+ +2CO;-+ 20 H-= U(OH)2(COJ3- 4 I.33 22

Binary parameters

Species p(o) p(l) @) C+ Ref.

tJ4+-cl- 1.644 15.5 0.00 0.0995UOH3+-CI-

201.0 7.856 0.00 0.00 20

usoi+–cl- 1.64 0.00 0.00 –0.2635 21

Na+–U(COJ$- 1.5 31.3 0.00 0.00 21K+–U(COJ!- 1.5 31.3 0.00 0.00 22


u(so’J>-cl- 0.29 21

model parametrization activities also have been conducted for U(IV) withCl-, SO1-, and CO$- (Table V). No model parameter has been publishedexplicitly for Np(IV) or Pu(IV) species, although such investigations arecurrently being conducted by the authors at PNNL. Values of the Pitzerion–interaction parameters for U(IV) and Th(IV) speeies appear to correlatewell and it is likely that will hold true for the Np(IV) and Pu(IV) species aswell. It should be pointed out that often there is a redundancy betweim thecalculated standard state equilibrium constant and the Pitzer ion–interactionparameters, especially when data at high electrolyte concentration are usedin the model parameterization. Therefore, in developing the Pitzer ion-interaction parameters for U(IV) species, the Th(lV) values were often usedas a guide or were fixed at the Th(IV) values. This co+uldhave influencedthe numerical closeness of the parameters. Nevertheless, the large. magnitudeof the values of these parameters, particularly for the highly charged electro-lyte types, clearly demonstrates the necessity for accurately determining thesemodel parameters.

3.3. Pentavalent Actinides

Aqueous species of Np(V) and Pu(V) are the principal pentavalentactinides of concern. Of these, Np(V) is stable over a broader range of pH

.

- .——.. —— --.---—

.,:.... ..

.

>’.

-..:..,

.

546 Felmy and Rai

-1-1 Calculated 4.67m NrtCl(no mixhg)

!& -------- =

●.- ..-.* *.-2 ..09

-m● -*-

t●

●

4 ● 0 ““k? 0 Calculated 2.33m NaCl (no mixing) .

●✏

9_-

E -4- .* /gc+g-5-A

New Model (4.67m NaCt)

-6

Calculated 2.33m NaCl (C104-terms)-7- 1 ----- ---

-m . . . . . .- .--.= .

Calculated 4.67m NaCl (C104-terms)-a

o 0.5 1 1:5 2 isNazC03 (M)

Fig. 7. ThOz(am) volubility data in mixed NajC03 and NaCl solutions. Experimental data(open symbols represent 2.33 m NaCl, closed symbols represent 4.67 m NaCl).

and redox conditions in aqueous solution than is Pu(V). Hence, the onlyPitzer ion-interaction parameters available for pentavalent actinides are forNp(V) species. To the best of our knowledge, there are no known valuesavailable for Pu(V) species, although recent unpublished work “at PNNLindicates that the values for NpO~–Cl - serve as useful analogs for thecorresponding PuO~ parameters. Unfortunately, ion–interaction parametersand solution-phase equilibrium constants for NpOJ species, which would beused to model the aqueous thermodynamics of Np(V) to high ionic strength,are available only for Cl-, CO;-, &d selected organic chelates (Table VI).

The development of the model paritmeters ‘summarized in Table VI is -the result of efforts primarily at SNL, LBL, LANL; and Institut fiir Radio-chemie. These studies began with the work of Novak and Roberts}25J Necket af. ,(29) and Fanghanel et al.(m) all in 1995. These onginal studies showed -quite similar trends in terms of the parameters that were required to describethe experimental data and in terms of the absolute magnitude of the modelingparameters. Later, Runde et a/.c2b)presented a somewhat modified model ofthe same chemical system (i.e. Pu(V) species in Na-C03–HC03–Cl–CI01–H–OH–HZO solutions at 25°C) as well as comparisons of their model predic-

.

..

... ..

,.

Aqueous Thermodynamics of Actinide Speeies 547

Table VI. Pitzer Ion-Interaction Parametem and Log ~ (25”C) for Np(V) Species

Reaction Log k-’ Ref.

NpO~ + H20 = Np020H + H+ –11.31 30NpOJ -1-2H20 = NpOz(OH); + 2H+ –23.54 30NpO; + CO;- = Np02C0-; 5.03 30NpOJ + 2COj- = Np(COJ]- 6.47 30NpO: + 2CO:- = Np(COJ;- 5.37 30N@: + EDTA4- = NpOzEDTA~- 8.54 34NpO; + HEDTA3- = NpOzHEDTA2- 4.98 34”N@; i- HJEDTA2- = Np02H2EDTA- 3.42 34

Binary parameters

Species B(o) $(1) p(z) c+ Ref.

NpO~-Cl- 0.1415 0.281 0.00 0.00 30NpOzEDTA’--Na+ 0.6830 0.5911 0.00 0.00 34NpOzHEDTA2--Na+ 0.4733 – 1.504 O.CQ O.CQ 34NpOzH2EDTA–-Na+ –0.8258 0.2575 0.00 0.256 34NpOzCO;–Na+ 0.10 0.34 0.00 0.00 30NpOz(COJ~--Na+ 0.48 4.4 0.00 0.00 30NpOz(COJj--Na+ 1.8 22.7 0.00 0.00 30


NpOJOH)j-Cl- -0.24 30Np02CO~-Cl- –0.21 30NpOz(COJ~--Cl- -0.26 30NpOz(CO&--CL- –0.26 30CO~--Np02(CO&- -1.9 27Np02(OH)-Cl- –0.19 30

.,..-...

..

. .

. ..

tions with those of Novak and RobertsJ”) Later Novak et a/.t27J and, veryrecently, Al Mahamid et uL@) have accepted many of the earlier valuesdetermined by the Institut fiir Radiochemie group and have applied thesevalues to similar chemical systems containing K+ with reasonable success.

‘, These parameter values are listed in Table VI. Ion-interaction parameters forNp(v) species with EDTA determined at Flordiaincluded in Table VI.

3.4. Hexavalent Actinides

In the case of hexavalent actinide species,

State University(~) are also

U(VI) and Pu(VI), limitedion-interaction parameters are available- only for U(W) (Table VII). Noparameters are available exclusively for PuO$+, although, based on structural

548 Felmy and Rai

.

Table VII. Pitzer Ion-Interaction Parameters and Log ~ (25°C) for U(W)

Reaction Log K- Ref.

UO;+ + EDTA4- = U02EDTA2- 13.16 34UO;+ + HEDTA3- = UOzHEDTA-EDTA4- + H+ = HEDTA3-

8.34 3410.57 34 ,

Binary parameters

Species p(m) @l) 9(2) C+ Ref.

uo~+-cl- 0.4274 I .644 0.00 -0.0184 44Uo;+-soi+ 0.322 1.827 0.00 -0.0176 44Na+-UOzEDTA2- –0.1516 1.74 0.00 0.095 34Na+-UOzHEDTA- 0.382 0.257 0.00 0.172 34

considerations, the parameters for UO~+ are expected to be good analogs forthe PuO~+ values.

One of the principal reasons for the limited availability of ion–interactionparameters for U(W) species relevant to natural waters is the presence ofseveral mononuclear and polynuclear hydrolysis species in aqueous solutionin the neutral or near-neutral pH region!37) The possible presence of thesehydrolysis species has made it difficult to develop Pitzer ion–interactionmodels for hexavalent actinides outside of the stability region of the uranylion, because it is difficult to attribute observed changes in thermodynamicdata (EMF, volubility, solvent extraction) to individual ion-interactions orcomplexation reactions. This situation has resulted in recent attempts toestimate the Pitzer parameters based on correlations with charge type, differ-ences in charges for reactions, and other variables!~s) In addition, the use ofmolecular modeling approaches are currently being applied at PNNL to betterestimate standard state equilibrium constants. These values are then used toreinterpret experimental data to evaluate Pitzer ion–interaction parameters,

—

as has been demonstrated for alkaline earth cations.@)Hence, in the c;se .of hexavalent actinides, the presence of multiple

species in solution at neutral to near-neutral pH conditions has made itdifflcr.dt to develop accurate ion–interaction models that are valid to high .ionic strength. This is unfortunate because this pH region is also the regionmost relevant to natural waters. Clearly more research is still needed inthis area.

4. CONCLUSIONS

Application of the Pitzer ion–interaction approach to the study of actinideaqueous species has proved to be a challenging test because of the large

.

—.—.

, .


number of chemical species in solution and the wide range of electrolytetypes involved. Fortunately, a significant database of model parameters hasbeen developed, especially for the trivalent and tetravalent actinides, and canbe used to model the aqueous thermodynamics of actinide species to highelectrolyte concentration. However, much work needs to be done on thepentavalent [especially Pu(V)] and hexavalent actinides in near-neutral pHconditions as well as the trivalent and tetravalent actinides under extremeconditions, such as high base concentration. In addition, reliable data are notavailable for actinides at higher than ambient temperature. However, bwauseof the progress that has been made on developing accurate modeling parame-ters at 25°C, it should now be possible to extend these models to highertemperatures if appropriate experimental data become available.

ACKNOWLEDGMENT

This manuscript is dedicated to the memory of Professor Kenneth S.Pitzer whose personal help and encouragement contributed immeasurably tomany of the studies cited in this work.

This research was conducted at Pacific Northwest National Laboratory,operated by Battelle for the U.S. Department of Energy. This research wassuppoied by the following projects: the Japan Nuclear Cycle DevelopmentInstitute, the U.S. Department of Energy Environmental Management Sci-ences Program (project #26753), and Sandia National Laborato~’s ActinideVolubility-Source “Term Project (AT-8746).

APPENDIX >.

These appendix tables include Pitzer ion–interaction parameters used — -in the development and parameterization of the ion-interaction parametersfor actinide species. ,

. .Table A.L Binary Ion-interaction Parameters for Major Electrolyte Ions

Speeies @o) p) p(z) 0 Ref.

Na+-Cl- 0.0765 0.2644 0.00 0.00127 2Na+-SO~ 0.01958 1.113 0.00 0.00497 2Na+-HSO~ 0.0454 0.398 0.00 0.00 2Na+-OH- 0.0864 0.2.S3 0.00 0.0044 2Na+-HCO~ 0.0277 0.0411 0.00 0.00 2Na+-CO~- 0.0399 1.389 0.00 0.0044 2K+–Cl- 0.04835 ().2122 0.00 –0.00084 2K+–SO~ 0.04995 0.7793 0.00 0.00 2

. ..... ..-

.,:,

,..+

550 Felmy and Rai

Table A.I. Continued.

Species p(o) p(l) B(2) @ Ref.

K+-HSO;K+-OH-

K+-HCO~K+-COj-@+-cl-

Ca2+–SO~Caz+-HSO~Ca2+-OH-Ca*+–HCO:M~-Cl -Mga+.S@

M&-HSO~Mg2+-HCO~M@ H+-CI-H+-CI-

H+-S~-H+-HSO~Na+-H$O~Na+–HPO~-Na+–POj-

–0.00030.12980.02960.14880.31590.200.2145

–0.17470.40.3S2350.22100.47460.329

–0.100.1775().()298

0.2065–0.0533–0.0583

0.17813

0.17350.320

–0.0131.431.6143.19732.53

–0.23032.9771.68153.3431.7290.60721.6580.29450.000.55560.03961.46553.8513

0.000.000.000.000.00

–54.24O.CK)

–5.72O.oa0.00

–37.230.000.000.000.000.000.000.000.000.00

0.000.0041

–0.008–0.0015–0.00034

0.000.000.000.000.005190.0250.000.000.000.00080.04380.000.007950.029380.05154

22222222222222222555

.

Table A.II. TernaryIon–Interaction Parameters for Major Electrolyte Ions”

Species Ternary parameter

Na+-K+f.Ja+-K+-cj-

Na+-K+-SO~-Na+-K+–HCO~Na+–K+–@-

Na+-Caz+Na+-Ca~+-Cl-

Na’-Ca2+-SO&Na+_MgZ+Na+-Mg2’–Cl-Na’-Mgz+-S@4-N&+-H+Na+-H+–Cl - .Na+-H+–HSO;K+_Ca2+

K+–Ca~+_C[-K+_Mg~+_Cl-

K+-Mg2+-SO~-K+-H+

–0.012–0.0018–0.010–0.003

0.0030.07

–0.007–0.055

0.07–0.012–0.0$5

0.036–0.004–0.0129

0.032–0.025–0.022–0.048

0.005

..—.

. .

—.——.——.—. -.—.:’.,.-.:,.,,-.:, . .

Aqueous Tbermodymamics of Actinide Species 551

,., .

Table A.If. Continued.

Species Ternary parameter

K+-H+-Cl- –0.01 iK+-H+–SO~- 0.197K+- H+-HSOZ –0.0265Ca~+-M$+ 0.007Caz+–Mgz+-Cl- –0.012&~+-Mg~+_S@- 0.024Ca2+_H+ 0.092CaZ+_H+-Cl- –0.015Mg2+-MgOH--Cl- 0.028Mg2+–H+ 0.10Mgz+–H+_Cl- –0.01 1Mg2+_H+_Hso: –0.0178cl-–so~- ().o~

C1--SO~--Na+ 0.0014cl--soj--ca~+ –0.018C1--SO~--Mgz+ –0.004CI--HSOI –0.006Cl--OH- –0.050C1-–OH--Na’ –0.006CI--HCO; 0.03CI-–HCO;–Na+ –0.015C1-–CO~--Na+ 0.0085

;. Cl-–CO:--K+ 0.004SO~--HSO;-Na+ –0.0094

..S@--HSO;-K+ –0.0677SO~-–HSO;-Mg2+ –0.0425s@--OH- –0.013S~--OH--Na+ –0.009SO~-_OH--K+ –0.050SO~--HCOI 0.01SO~-–HCO;-Na+ –0.005SC&-HCO-~_M#+ –0.161Soi--co;- - 0.02SO~-–CO~--Na+ –0.005

.+ S@--C&-K+ –0.009..-.,.

. . , aAll data from Ref. 2.-.-:..

REFERENCES

1. A. R. Felmy and J. H. Weare, Geochim. Comrochim. Acts 50, 2771 (1986).2. C. E. Harvie, N. Moller, and J. H. Weare, Geochim. Cosmochim. Ac[a 48, 723 (1984).3. K. S. Pitzer, ./. ,Wutimr Chemisuy 4,249 (1975).4. K. S. Pitzer and L. F. Silvester, ./. Phys. Chem. 82, 1239 (1978).5. K. S. Pitzer and G. Mayorg& J. Phys. Chem. 77, 2300 (1973).

: -.:.----

_.. , -—.

.:. .. . . ... .

.—

552 Felmy and Rai

6. K. S. Pitzer ~d G. Mayorg% J. Solution Chern. 3,539 (1974).7. A. R. Felmy, D. Rai, J. A. Schramke, and J. L. Ryan, Radiochim. Acts 48,29 (1989).8. A. R. Felmy, D. Rai, and R. W. Fulton,Radiochim. Ac-za50, 193 (1990).9. A. R. Felmy, D. Rai, M.’J. Mason, and R. W. FuIton, Radiochim. Acfa 69, 177 (1995).

10. D. Rai and A. R. Felmy, in:, Peq%rmance Assessment Center for Engineered Barriers(PACE) Program FY 1990 Repart, VO1.2 (Power Reactor and Nuclear Fuel DevelopmentCorporation (PNC), Japan, PNC PA0865 91-001, 1991).

11. D. Rai, A. R. Felmy, and R. W. Fulton, Radiochim. Acts 56,7 (1992).●

12. D. Rai, A. R. Felmy, M. J. Mason, and R. W. Fulton, in Perjiormance Assessment Centerfor Engineered Barriers (PACE) Program FY 1991 Report, Vol. 1 [Power Reactor andNuclear Fuel Development Corporation (PNC), Japan,PNCZA086592-001, 1992].

13. D. Rai, A. R. Felmy, and R. W. FuIton, J. Solution Chenrisrry 24,879 (1995).14. L. Rae, D. Rai, A. R. Felmy, R. W. Fulton, and C. F. Novak, Radiochim. Acta75, 141 (,1996).15. L. Rae, D. Rai, A. R. Felmy, and C. F. Novak, 212tb ACS Symp. Proc. Radimruclide

Speciarion in Real Systems (1996),”manuscript submitted. ,

16. A. R. Felmy and D. Rai, & Solution Chem. 21,407 (1992).17. A. R. Feimy, D. Rai, and M. J. Mason, Radiochim. Acts 55, 177(199 1).18. A. R. Felmy, D. Rai, and M. J. Mason, Radiochim. Acts 62, 133(1993).19. A. R. Felmy, D. Rai, S. M. Sterner, M. J. Mason, N. J. Hess, and S. D. Conradson, L

Saktion Chem. 26, 233 (1997).20. D. Rai, A. R. Felmy, S. M. Sterner, D. A. Moore, M. J. Mason, and C. F. Novak, Radiochim.

. ..-.,... Acra 79, 239 (1997)..... ... .. ... . .. . . .. . . . .’.... :j ..’ . .. ..

J-.. ..- .. .. .... ., ..’.’.”’ 21. D. Rai, L. Rae, H: Weger, A. R. Felmy, and G. R. Choppin, Thermodynamic Data for?.,. .,. .; .:”,.-.,-..: ,. .’.,-, :- .,., . .. .... Predicting Concentrations qf Th(lV), U(IV), IVp(IV),and Pu(IV) in Geologic Environments

PNWD-2428, Rev 1,(Pacific Northwest National Laboratory, Richland, Washington, 1998).22. D. Rai, A. R Felmy, N. J. Hess, D. A. Moore, and M. Yui, Radiachim. Acre. 82, 17 (1998).23. D. Rai, N. J. Hess, A, R. Felmy, D. A. Moore, and M. Yui, Radiochirn Acts, in press (1998).

24. D. Rai, L. Rae, and D.A. Moore, Radiochim Acra 83, (1998).25. C. F. Novak and K. E. Roberts, Mat. Res. SK Symp. Pm. 353, 1119 (1995).

26. W. Runde, M. P. Neu, and D. L. Clark, Geochim. Cosmochim. Acra 60, 2(@5 (1996).27. C. F. Novak, L Al Mahamid, K. A. Becraft, S. A Carpenter, N. Hakem, and T. Prussin,

J. Sokimr Chem. 26,681 (1997).28. I. Al Mahamid, C. F. Novak, K. A. BecrafL S. A. Carpenter, and N. Hakem, Radiochim. . . .

.+ Acts 81,931 (1998). -29. V. NeclG T. Fimghanel, G. Rudolph, and J. I. Kim, Radtichim. Acts 69,39 (1995).

--~

30. Th. Fanghanel, V. Neck, and J. L Kim, Radiochim. Acla 69, 169 (1995).~ 1. Th. Fanghanel and J. I. Kim, J. Alloys Comp. 271-273,728 (1998).32. Th. Fanghanel, H. T. Weger, Th. Konnecke, V. Neck, P. Pavi&Hartmann, E. Steinle, and

J. 1. Kim, Radiochim. Acza 82,47 (1998).----- . .. . .. . ---- . . . ... . .. . . . . . --, .-,.,..55.

34.

35.36.37.

38.

1n. ranghanel, H. 1. weger, u. scrnme% ana J. L KIIII, Kacfmctrrm. Ackz, &z, >> ( lY%). .

0. S. Pokrovsky, M. G. Bronikowski, R. C. Moore, and G. R. Choppin, Radiochim. Acfa80, 23 (1998).F. J. Millero, Geachim. Cosmochim. Acts, 56,3123 (1992).A. Plyasunov, T. Fanghanel, and 1. Grenrhe, Acts Chemica Stand. 52,250 (1998).I. Grenthe, J. Fuger, R. J. M. Konnings, R. J. LeMire, A. B. Muller, C. Nguyen-Tnmg,and H. Wanner, Chemical Thermodynamics of Uranium. Nucle~ Energy Agency Organiza-tion for Economic Co-operation and Development, Elsevier, Amsterdam (1992).A. R. Felmy, D. A. Dixon, J. R. Rustad, M. J. Mason, and L. M. Onishi, J. Chem.Thertrmdyn. 30, 1103 (1998).

.

- -----

..

,.


39. D. Rai, L. Rae, H. Weger, A. R. Felmy, and G. R. Choppin, 7kermodynamic Data for.Predicting Concenlrazions of Pu(III), Am(III), and Cm(III) in Geologic EnvironmentsPNWD-2427, Rev 1(Pacific Northwest National Laboratory, Richk+nd,Washington, 1998a).

40. R. J. Silva, The Solubiliries of Crystalline Neodymium and Americium Trihydmxides, LBL-15055 (Lawrence Berkeley Laboratory, Berkeley, Califomi% 1987).

41. D. Rai, R. G. Strickert, D. A. Moore, and J. L. Ryan, Radiochim. Acts 33,201 (1983).42. W. J. McDowell and C. F. Coleman, ~. {norg. Nucl. Chem. 34,2837 (1972).43. D. Rai, A. R. Felmy, D. A. Moore, and M. J. Mason, Mat. Res. Sot. Symp, Proc. 353,

1143 (1995).44. K. S. Pitzer, Activity Coejjkients in Electrolyte Solutions, 2nd edn. CRC Press (Boca

Raton, ~ 199l).45. R. J. Sifva, G. Bidog[io, M. H. Rand, P. B. Robouch, H. Wanner, and I. Puigdomenech,

Chemical Thermodyns 2: Chemical Thermodynamics of Americium (Elsevier, Amster-dam, 1995).

46. R.N. Roy, K. M. Vogel, C. E. Good, W. B. Davis, L. N. Roy, D.A. Johnson, A. R. Felmy,and K. S. Pitzer, ~. Phys. Chem. 96, 11,065 (1992).

47. J. L. Ryan and D. Rai, lnorg. Chem. 26,4140 (1987).48. D. Rai, A. R. Felmy, and J. L. Ryan, lnorg. Chem 29,260 (1990).49. K. S. Pitzer and L. F. Sitvester, J. Solution Chem. 5, 269 (1976).50. L. Rae, D. Rai, and A. R. Felmy, Radiochim Acts 72, 151-155.

---- .--- .

- . ...-.

Documents

Application of Pitzer’s Equations for Modeling the Aqueous Thermodynamics … · Aqueous Thermodynamics of Actinide Species in Natural Waters: A Review~ n Andrew R. Felmy* and Dhanpat